Content recommendation model training method and device, server and storage medium

• Main Authors: LI HAO, YANG PINCI
• Format: Patent
• Language: Chinese; English
• Published: 20.12.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to the technical field of artificial intelligence, in particular to a training method and device of a content recommendation model, a server and a storage medium. The method comprises the steps that a training data set is obtained, the training data set comprises user features of a sample user and sample content, the sample content has multiple modals, and then the user features of the sample user and the sample content are obtained on the basis of preset time use proportions for a single-flow sub-model and a multi-flow sub-model; and alternately inputting the user characteristics and the sample contents into the single-stream sub-model and the multi-stream sub-model to obtain recommended contents for the sample contents output by the content recommendation model, and training the content recommendation model based on the recommended contents to obtain a target content recommendation model. By adopting the method provided by the invention, multiple modes can be aligned in different seman